OSMnx package provides the possibility to obtain a networkx-graph representation of a street-network from OpenStreetMap with a single line of code.Ī relevant section for our scenario can be obtained by specifying center and distance for osm-nodes to be included: CENTER = (53.55668, 9.92815) OSMnx/networkx for retrieving geo-information and calculating shortest-paths.The simulation model for the simplistic scenario is created with: The tour starts & ends at the patisserie: Zentrales Fundbüro (lost-and-found office)įor the sake of simplicity, the order in which nodes are visited is assumed to be fixed. Monkeys Music Club, and the publishing house The simulation model of this post will be based on a mini-case-study.įacing the fact that humanity gets more and more used to home-delivery of even everyday necessities, the small French bakeryĭie Patisserie in Hamburg-Altona wants to deliver sweet pastries to nearby customers.ģ major purchasers were identified: the lost-and-found office Obviously, those commercial features come at a price, so let’s see whether this kind of model can also be realized with different means. The data-driven simulation model allows to quickly calculate KPIs for various transshipment node locations and different types of transport equipment in a multi-tier supply-chain network.įollowing the success of the feature, Anylogic even builtĪnyLogistix, combining pre-built & customizable simulation models with a commercial solver for integrated supply chain planning & optimization. Previous project used Anylogic to evaluate different designs and delivery schemes for a more sustainable urban logistics network in the city center of Grenoble in France. This is cool because it can be used to simulate entire supply chains, including means to provide a great, tangible visualization for complex problems. It allows to place elements on a map and move them along existing routes, based on real spatial information.
A really nice and quite unique feature ofĪnylogic is the possibility to include GIS-maps into simulation models.
0 kommentar(er)
